Intranuclear HSV-1 DNA ejection induces major mechanical transformations suggesting mechanoprotection of nucleus integrity

Figures

• 

  Fig. 1.. Imaging of a reconstituted capsid–nuclei system confirms specific binding of A- and C-capsids to the NPCs at the nuclear membrane. Confocal fluorescence microscopy images show binding of GFP-HSV-1 A- and C-capsids (green) to DAPI-stained isolated nuclei (blue), in the presence of cytosol and ATP-regeneration system. The addition of WGA decreases capsid binding to nuclei for both capsid types, which demonstrates that capsids bind specifically to NPCs as opposed to binding anywhere on the nuclear membrane. The images at the bottom of row two are a zoom-in of the individual nuclei.

• 

  Fig. 2.. Representative SR-SIM images show GFP-HSV-1 A- and C-capsids (green) bound to isolated rat liver nuclei (blue DAPI stain) with similar nuclear surface occupancy. As a complementary approach, we also visualized capsid binding to NPCs by ultrathin-sectioning EM, which further confirmed docking of capsids to the NPCs. Figure shows EM micrographs of A- and C-capsids bound to the NPC basket on a nucleus surface (the observed density inside the C-capsids corresponds to packaged DNA before it is ejected into the nucleus since sample was kept at 4 °C prior to fixation, while DNA ejection occurs at 37 °C; Ref. 10).

• 

  Fig. 3.. Quantification of capsids docked at the nuclear membrane. (A) Infection progression of GFP/HSV-1–infected Vero cells. At 3.5 hpi, the maximum number of capsids is reached before new capsids are assembled in the nucleus. (B) Steps of confocal fluorescence image analysis are shown for quantification of capsids per nucleus in infected Vero cells at 3.5 hpi. (C) Steps of confocal fluorescence image analysis are shown for quantification of capsids per isolated cell nucleus. First row is a 2D slice (one focal plane) from the 3D stack GFP capsids (green) on DAPI-stained nucleus (blue) in the second row. Next row is a Z-projection in 2D of all confocal plane slices in the 3D stack that was used for quantification of individual capsids, built using nonoverlapping green pixels with the highest intensity from each slice (corresponding to capsids). Z-projection shows both channels and exhibits the projection of all capsids on a nucleus. The green channel was converted to a binary image using a threshold, as shown in the last row, in order to accurately compute number of capsids bound at each nucleus. Capsids outside of the nuclear membrane surface are not counted (as seen in the threshold…

• 

  Fig. 4.. A panel showing AFM FFM imaging sequence. (A) Zoomed-out top view optical image and (B) zoomed-in optical image before moving the cantilever above the nucleus. (C–F) A representative 3D topography AFM image of the nuclear membrane surface with well-resolved NPC baskets with HSV-1 A-capsids attached (similar images were obtained for C-capsids). Capsids were incubated with nuclei at 37 °C for 40 min in cytosol and ATP regeneration system. After incubation, nuclei were fixated with 2% GA, increasing nuclear envelope rigidity and improving AFM imaging resolution, and washed in CBB for imaging. AFM image shows that the nucleus surface of rat liver cells is densely covered with protruding NPC baskets, providing binding sites for HSV-1 capsids. Capsids on the nucleus surface are displaying faceted icosahedral features as well as resolved individual capsomer subunits characteristic of HSV-1 capsid structure. The capsid height measured on top of the nuclear membrane surface is ∼124 nm, which is in agreement with a cryo-EM–obtained value of 125 nm for HSV-1 capsid diameter (56).

• 

  Fig. 5.. (A) A representative 10 × 10 force volume map for unfixed nucleus with attached C-capsids (corresponding to 100 Fz curves) acquired over a 200 nm × 200 nm area at the center of each nucleus. (B) A representative Fz curve, selected from one of the acquired force volume maps: cantilever approach and indentation into C-capsid bound nucleus (solid red line). Some Fz curves displayed a “notch” in the deformation behavior (blue line). In this force curve, the cantilever first pushes into the capsid and then continues to indent into the nucleus. The observed drop in the Fz curve could reflect detachment or breaking of the capsid at the nucleus surface. When force indentation curves are acquired on nuclei without capsids, this notch is not detected. Force indentation curve acquired on a nucleus shows only one slope, which represents the response of the nucleus only. Moduli was calculated using the Hertz contact mechanics model. Contact point, depth, and force offsets were autoselected by the software. Fit range was adjusted as needed to maintain an indentation depth of 300 nm. Cantilever retraction (dashed red and blue lines) and fit using Hertz mechanics model for Young’s modulus…

• 

  Fig. 6.. Young’s moduli for 100 Fz curves in each force volume map were repeatedly collected for several nuclei as well as for several areas on the same nucleus. All moduli values were collectively plotted for each type of sample in a histogram for nuclei without capsids, nuclei with attached A-capsids, and nuclei with attached C-capsids, respectively. (A) Histograms of Young’s moduli for all samples without fixation, reflecting chromatin stiffness. (B) Histograms of Young’s moduli for all samples with 2% GA fixation, reflecting relative nuclear lamina stiffness. The Gaussian fit to the data in the histograms yields the average value and the SE shown in Table 1.